© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 66550
    Prev
    Next
    INDEX
    Explanations

    exist in

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron activates on numeric tokens (numbers and numerals) in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     ſy
    -2.19
     westfalen
    -2.17
    釤
    -2.16
     filt
    -2.11
     ſta
    -2.09
     kling
    -2.09
     kram
    -2.09
     herren
    -2.08
     calo
    -2.06
    笵
    -2.03
    POSITIVE LOGITS
     There
    2.08
     exist
    1.99
     for
    1.99
     on
    1.94
     in
    1.87
     Each
    1.82
    {
    1.80
     Existe
    1.78
    ↵
    1.73
    There
    1.73
    Activations Density 0.038%

    No Known Activations